Understanding and generating spatially coherent layouts from natural language remains a fundamental yet challenging task for large language models (LLMs). Existing LLMs often struggle to capture explicit geometric relationships and structural dependencies between objects. To address this issue, we propose SG-Layout, a graph-guided layout generation framework that explicitly incorporates structured spatial knowledge into LLMs. SG-Layout follows a two-stage training paradigm: (1) a graph-language feature alignment stage, where a relational graph encoder and a projector are trained to map scene-graph embeddings into the LLM's linguistic space; and (2) an instruction tuning stage, where LoRA-based adapters enable efficient fine-tuning for instruction-driven layout generation while keeping the backbone frozen. We evaluate SG-Layout on image layout generation, indoor scene synthesis and robotic object rearrangement tasks. Experimental results show that SG-Layout improves spatial reasoning accuracy and geometric consistency over the compact open-source backbone, with particularly clear advantages in relation-dense and compositionally complex scenes. These results highlight the effectiveness of graph-structured feature alignment for enhancing controllable layout generation.

Mobile manipulation is a fundamental capability in embodied intelligence robotics. The growing demand for robust and generalizable manipulation in unstructured household environments has driven rapid progress in embodied intelligence platforms. However, achieving a seamless transfer across the real-to-sim-to-real cycle faces three key challenges, including costly high-fidelity simulation scenes reconstruction, the complexity of systematic strategy evaluation in simulation, and incompatible real-world deployments. To address these challenges, we develop BestMan, a scalable and seamless real-to-sim-to-real platform that bridges the gap between the simulation and the real world, enabling effective strategy development, integration, and deployment for household mobile manipulation. Specifically, we design a novel Automated Scene Generation (ASG) module to reconstruct realistic simulations from real observations. Then, we propose a simulation-guided task formalization and skill learning architecture that supports the flexible integration and large-scale evaluations of hybrid skill strategies in simulation. Finally, to enhance the real-world scalability, we develop a Hardware-agnostic and Unified Middleware (HUM) to ensure seamless and compatible sim-to-real transfer across heterogeneous mobile manipulators for real deployments. Experimental results demonstrate the superior performance of our proposed platform in establishing standardized benchmarks and facilitating promising research in the field of mobile manipulation.

Route recommendation systems commonly adopt a multi-stage pipeline involving fine-ranking and re-ranking to produce high-quality ordered recommendations. However, this paradigm faces three critical limitations. First, there is a misalignment between offline training objectives and online metrics. Offline gains do not necessarily translate to online improvements. Actual performance must be validated through A/B testing, which may potentially compromise the user experience. Second, redundancy elimination relies on rigid, handcrafted rules that lack adaptability to the high variance in user intent and the unstructured complexity of real-world scenarios. Third, the strict separation between fine-ranking and re-ranking stages leads to sub-optimal performance. Since each module is optimized in isolation, the fine-ranking stage remains oblivious to the list-level objectives (e.g., diversity) targeted by the re-ranker, thereby preventing the system from achieving a jointly optimized global optimum. To overcome these intertwined challenges, we propose SCASRec (Self-Correcting and Auto-Stopping Recommendation), a unified generative framework that integrates ranking and redundancy elimination into a single end-to-end process. SCASRec introduces a stepwise corrective reward (SCR) to guide list-wise refinement by focusing on hard samples, and employs a learnable End-of-Recommendation (EOR) token to terminate generation adaptively when no further improvement is expected. Experiments on two large-scale, open-sourced route recommendation datasets demonstrate that SCASRec establishes an SOTA in offline and online settings. SCASRec has been fully deployed in a real-world navigation app, demonstrating its effectiveness.

Trajectory computing, which encompasses trajectory data management and mining, has been widely studied owing to its role in practical applications such as location services, urban traffic, and public safety. Traditional methods, focusing on simplistic spatio-temporal features, face challenges of complex calculations, limited scalability, and inadequate adaptability to real-world complexities. In this paper, we present a comprehensive review of the development and recent advances in trajectory computing, from deep learning to the more recent large language models. We first define trajectory data and provide a brief overview of widely-used deep learning models. Systematically, we explore deep learning applications in trajectory management (pre-processing, storage, analysis, and visualization) and mining (trajectory-related forecasting, trajectory-related recommendation, trajectory classification, travel time estimation, anomaly detection, and mobility generation). Furthermore, we discuss emerging research directions and recent advancements in large models (represented by foundation models and large language models) for trajectory computing, which promise to reshape the next generation of trajectory computing. We also provide practical guidance on method selection for each task and analyze the cost–accuracy trade-offs of large models against specialized deep models. Additionally, we summarize application scenarios, public datasets, and toolkits. Finally, we outline current challenges in trajectory computing research and propose future directions. Relevant papers and open-source resources have been collated and are continuously

Safe and precise maneuvering of quadrotor unmanned aerial vehicles (UAVs) in high-speed wind environments remains a critical challenge. Wind disturbances are nonlinear, time-varying, and difficult to model, causing traditional controllers to struggle with perception and compensation, especially under unseen wind distributions. To address these limitations, we introduce WA-TD3, a data-driven control framework that enables real-time wind disturbance perception and adaptive compensation without dedicated wind sensors. WA-TD3 employs a deep residual network to extract wind characteristics from temporal patterns in state deviations, forming a dynamics residual-driven perception mechanism that implicitly models and compensates for unknown winds. This residual is integrated into a perception-augmented reinforcement learning architecture, providing the policy with enhanced state information for proactive disturbance-aware control. Extensive experiments on complex trajectories under varying wind intensities demonstrate that WA-TD3 consistently outperforms state-of-the-art methods, achieving over 62% improvement in tracking accuracy under strong winds.

Software security has been a long-standing and prominent topic in both industry and academia. However, with the increasing deployment of smart devices across various architectures, there is now a significant demand for cross-architecture software. For instance, the Heartbleed vulnerability (CVE-2014-0160), classified as CWE-125 (Out-of-bounds Read) and disclosed in the OpenSSL library in 2014, serves as a prime example of a widely-deployed software vulnerability capable of affecting systems across diverse computing environments due to OpenSSL’s broad adoption. More broadly, memory-related vulnerabilities remain a persistent and significant threat, accounting for a substantial proportion of reported CVEs in recent years. Moreover, performing static cross-architecture vulnerability analysis on binary code poses particular challenges due to the variations in instruction sets across different architectures. In this paper, we present a novel static approach, called BinCros, to identify known vulnerabilities across different optimization levels and CPU architectures in binary code. It mainly consists of three phases: 1) Generating ground-truth intermediate representation (IR) data from binary code based on domain knowledge. To address the cross-architecture problem, we consider IR, which is a set of smaller, lower-level instructions that break down, describe, and capture all the effects of an instruction from standard architectures like x86, ARM, and MIPS; 2) Learning good embeddings based on the cross-architecture ground-truth datasets. We employ a code-code learning method to capture the semantic differences and maximize the distribution divergences between vulnerable and non-vulnerable samples; 3) Building a prediction model based on high-level feature representations. To demonstrate the effectiveness of BinCros, we conducted experiments with a series of baselines including single architecture-based techniques, cross-architecture-based techniques, and so on. Experimental results show that BinCros outperforms the baselines by at least 15% in terms of F1-measure in almost all cases. We believe this work will inspire other researchers to consider natural language processing-related techniques for cross-architecture software vulnerability detection.

Task-vector–based model merging enables low-cost, training-free multi-task learning for large language models, but suffers from severe performance degradation due to task conflict. Prior mitigation strategies largely rely on validation data for costly hyperparameter tuning, limiting both interpretability and practicality. We therefore propose OPIC, an evolutionary optimization–based model merging framework. Our preliminary experiments reveal that the degradation of In-Context Learning (ICL) capabilities is a primary driver of task conflict. Motivated by this insight, we formulate model merging as an optimization problem with ICL preservation as the objective. OPIC introduces a hierarchical refinement operators and optimizes it using self-generated data, effectively eliminating the reliance on external validation sets. Experimental results demonstrate that OPIC achieves an average performance retention of 80.73%, outperforming SOTA methods and improving by up to 11.1% over recent validation-free approaches. In addition, OPIC is compatible with existing merging pipelines, offering a new alternative solution for deploying without validation dependencies. Code is available at: https://anonymous.4open.science/r/OPIC-CFFE.

Base placement is particularly critical for navigation manipulation tasks in environments, where improper placement can severely hinder task execution if the object’s kinematics are not properly taken into account. In this work, we present MoMa-Pos, a framework that determines the base placement for mobile manipulators in such environments. MoMa-Pos leverages a graph-based neural network to predict object importance and selectively reconstructs the environment by prioritizing task-relevant key objects, enhancing computational efficiency and ensuring that only essential kinematic details are processed. Moreover, MoMa-Pos integrates inverse reachability maps with environmental kinematic properties to determine feasible base placement tailored to the specific robot model. Extensive evaluations demonstrate that MoMa-Pos outperforms existing methods in both real and simulated environments, offering improved efficiency, precision, and adaptability across diverse settings and robot models.

Automated guided vehicle (AGV) scheduling problem has become a key challenge in workshop logistics system. This study addresses the integrated scheduling problem of AGV task assignment and path planning in matrix manufacturing workshop. Three typical path collision scenarios are considered: same-direction collision, opposite-direction collision and node collision. A mixed integer linear programming model is formulated with delivery time deviation penalties, travel expenses, AGV deployment costs, and collision fines as objective. To solve the AGV scheduling problem (AGVSP), an improved memetic algorithm (IMA) is proposed. An innovative three-layer encoding-decoding method is designed. This method represents the AGV path through segmented trajectories to achieve efficient collision detection. Then, a scanning heuristic is proposed to construct highquality solution. And a dynamic backtracking mechanism is developed to resolve collisions through local path reconstruction. Simulation experiments demonstrate that the proposed approach ensures operational safety while achieving optimal scheduling solution for the workshop.

Aiming at the nonstationarity and complex spatio-temporal dependence modeling problems caused by sudden anomalies in traffic flow prediction, this paper proposes a segmented attention and anomaly-aware spatio-temporal network (SA2ST-Net), which can decouple the modeling anomaly characteristics while capturing the normal traffic flow law. The model includes four modules: the adaptive spatio-temporal graph construction module provides dynamic road network topology, the segmented attention gated recurrence module captures long-range spatio-temporal dependence, the anomaly-aware spatio-temporal attention module mines the abnormal spatio-temporal propagation mode through the gating mechanism, and the transformer-based anomaly detection module captures abnormal signals and guides the model to pay more attention to abnormal areas with the help of loss function. Experiments on four real data sets show that SA2ST-Net is significantly superior to the existing advanced methods in multi-step and abnormal scene prediction, taking into account the prediction accuracy and computational efficiency.

Articulated object manipulation is a unique challenge for service robots. Existing methods employ end-to-end policy learning, visionmotion planning, and large-language/visual-language model (LLM/VLM), but often overlook the diversity of articulated objects and the complexity of interactions between end-effector and handle, leading to limited generalization and destructive collisions. To address this, we propose GSAM, a generalizable and safe robotic framework for articulated object manipulation. Specifically, a vision-based perceiver generates the kinematic parameters. Considering that pre-trained markers in perceiver yield raw estimations that may deviate from commonsense, we present a f ine-tuned VLM-based refiner, using chain-of-thought (COT) commonsense reasoning to refine perception. To prevent destructive collisions, we design an interaction constraint function generator, integrating articulated object, interaction pose, and obstacle avoidance knowledge into a base. LLM then functionalize these constraints and apply them to trajectory and posture planning. A kinematic-aware manipulation planner verifies reachability for trajectory and posture. Experiments on 50 hinge tasks across 5 object categories and 50 randomly initialized end-effectorhandle configurations show that GSAM reduces standard deviation by 3.1

Graph neural networks (GNNs) have been widely applied in software-defined network (SDN) to enhance network modeling and performance forecasting. However, the closed-box nature of deep learning makes GNNs difficult to interpret, hindering their broad use and the application of GNN-based SDN systems in engineering. In this paper, we propose a novel interpretation framework named GEEK-Explainer, designed to efficiently provide instance-level interpretation of GNNs in SDN. Specifically, we introduce a KernelSHAP-based scoring module to generate intuitive and human-friendly explanations for each performance prediction. To address conflicts in computation cost, we propose a soft discrete mask matrix that identifies a critical set of important nodes. Extensive experiments demonstrate that the RouteNet model can effectively learn the relationships among features, which can provide a better understanding of the prediction process with less computation cost. These findings improve the transparency and robustness of the model and promote the application of GNN-based SDN systems in engineering practice.

Dual-arm manipulation can improve throughput via parallel execution, but collecting bimanual demonstrations for training is costly and difficult. We present ExS2D, a hierarchical action expansion framework that enables dual-arm manipulation from single-arm supervision. ExS2D first generates structured subtasks from textual instructions while explicitly capturing temporal precedence. It then grounds each subtask into executable actions through subtask-guided action mapping in observation. Finally, precedence-aware action allocation and synchronized planning are performed by a multimodal large language model driven coordinator to select collision-free dual-arm executions. Simulation experiments demonstrate that ExS2D reduces the average execution steps by 54.4

Accurate prediction of slope displacement is an important prerequisite for building an effective geological hazard early warning system for disaster prevention and reduction. However, the inherent nonlinearity and time-varying characteristics of slope displacement evolution greatly affect the prediction accuracy. To improve the slope displacement prediction accuracy, a multi-modal data-driven Bayesian-optimized Convolutional Neural Network and Long Short-Term Memory (Bayes-CNN-LSTM) model was constructed. The performance of the model was evaluated using multi-modal monitoring data from the GuShan mine slope. Experimental results showed that the Bayes-CNN-LSTM model achieved an average coefficient of determination (R2) of 0.971, with a mean absolute error (MAE) of 0.444 mm and a root mean square error (RMSE) of 0.618 mm. Compared with the CNN-LSTM, LSTM, CNN, SVM, TCN, and Transformer models, the MAE of the constructed model was decreased by 25.1%, 31.3%, 32.3%, 24.1%, 24.7%, and 17.7%, respectively, and the RMSE decreased by 20.1%, 26.9%, 29.5%, 18.0%, 20.7%, and 12.4%, respectively. Furthermore, the proper integration of multi-modal data can effectively improve the prediction accuracy when extrapolating slope displacement. Based on rainfall and earth pressure data, the average MAE and RMSE of extrapolation (24-h) prediction using the constructed model were decreased by 30.2% and 24.6%, respectively. The model effectively improves the accuracy of slope displacement prediction and enhances the practicality of the slope safety monitoring system, providing valuable reference for slope safety monitoring.

Multi-turn dialogues and context-intensive tasks challenge Large Language Models (LLMs) to integrate long histories without sacrificing generation quality. Although prefix LLMs can better exploit historical context via bidirectional attention on prefix tokens, they are rarely used in practice because multi-turn training requires many duplicated triplets, and its bidirectional prefix prevents KV-cache reuse at inference time, driving up high cost and latency. To retain the contextual understanding of prefix mask while preserving the inference-time efficiency of causal mask, we introduce Intermittent Semi-working Mask (ISM), a masking scheme that injects sparse bidirectional attention into the causal backbone. ISM alternates bidirectional attention over query segments with unidirectional attention over answer segments, enabling the synthesis of in-context while preserving global causality. This design eliminates triplet expansion during training and maintains KV-cache reuse during inference, yielding latency comparable to standard causal LLMs. ISM is architecture-agnostic and parameter-free, adding only minimal latency. Across extensive evaluations, ISM outperforms causal baselines not only on multi-turn dialogue, but also on context-intensive tasks like mathematical reasoning.

Multi-label feature selection (FS) plays a vital role in multi-label learning as properly selected features can be used to substantively improve classification performance and reduce training time of a classifier. Existing multi-label FS methods primarily rely on the correlations among variables but neglect causality, and consequently lack interpretability. Although causal FS techniques based on Markov blankets have been widely investigated for single-label learning, its exploration for multi-label learning is rather limited because of the more complex causal relationships in multi-label data. In this paper, we present a multi-label causal FS method that leverages the introduced concept of unary approximate Markov blankets to identify causal structure of labels. Moreover, it combines the label-label, feature-feature and label-feature relationships in multi-label datasets and restores the features that are omitted due to equivalent information among features and labels. We conduct experiments on a variety of multi-label datasets and compare our proposed method with the state-of-the-art algorithms. The results show that our approach achieves significantly better performance in terms of a number of different metrics than them, thus greatly advancing the field of multi-label FS.

Existing industrial-scale navigation applications contend with massive road networks, typically employing two main categories of approaches for route planning. The first relies on precomputed road costs for optimal routing and heuristic algorithms for generating alternatives, while the second, generative methods, has recently gained significant attention. However, the former struggles with personalization and route diversity, while the latter fails to meet the efficiency requirements of large-scale real-time scenarios. To address these limitations, we propose GenMRP, a generative framework for multi-route planning. To ensure generation efficiency, GenMRP first introduces a skeleton-to-capillary approach that dynamically constructs a relevant sub-network significantly smaller than the full road network. Within this sub-network, routes are generated iteratively. The first iteration identifies the optimal route, while the subsequent ones generate alternatives that balance quality and diversity using the newly proposed correctional boosting approach. Each iteration incorporates road features, user historical sequences, and previously generated routes into a Link Cost Model to update road costs, followed by route generation using the Dijkstra algorithm. Extensive experiments show that GenMRP achieves state-of-the-art performance with high efficiency in both offline and online environments. To facilitate further research, we have publicly released the training and evaluation dataset. GenMRP has been fully deployed in a real-world navigation app, demonstrating its effectiveness and benefits.

Semantic segmentation is a fundamental task in computer vision with wide-ranging applications, including autonomous driving and robotics. While RGB-based methods have achieved strong performance with CNNs and Transformers, their effectiveness degrades under fast motion, low-light, or high dynamic range conditions due to limitations of frame cameras. Event cameras offer complementary advantages such as high temporal resolution and low latency, yet lack color and texture, making them insufficient on their own. To address this, recent research has explored multimodal fusion of RGB and event data; however, many existing approaches are computationally expensive and focus primarily on spatial fusion, neglecting the temporal dynamics inherent in event streams. In this work, we propose MambaSeg, a novel dual-branch semantic segmentation framework that employs parallel Mamba encoders to efficiently model RGB images and event streams. To reduce cross-modal ambiguity, we introduce the Dual-Dimensional Interaction Module (DDIM), comprising a Cross-Spatial Interaction Module (CSIM) and a Cross-Temporal Interaction Module (CTIM), which jointly perform fine-grained fusion along both spatial and temporal dimensions. This design improves cross-modal alignment, reduces ambiguity, and leverages the complementary properties of each modality. Extensive experiments on the DDD17 and DSEC datasets demonstrate that MambaSeg achieves state-of-the-art segmentation performance while significantly reducing computational cost, showcasing its promise for efficient, scalable, and robust multimodal perception.

Serverless computing offers fine-grained billing and elastic scalability, making it appealing for workflow execution. However, it also suffers from cold-start latency and stricter execution constraints. In contrast, traditional serverful cloud resources, such as virtual machines, provide coarser provisioning granularity but benefit from relatively lower unit cost. This work explores the potential of blending serverful and serverless resources to harness their complementary strengths for cost-effective workflow execution. We propose a hybrid resource management framework that dynamically allocates workflow tasks across both types of resources. A Budget-constrained Workflow scheduling algorithm for Blended cloud (BWB) is developed to minimize makespan while respecting user-specified budget. Evaluation experiments are conducted under real-world cloud settings by using realistic workflow applications. BWB is compared against state-of-the-art approaches for serverful, serverless, and blended clouds. Experimental results show that BWB consistently outperforms its state-of-the-art peers, achieving makespan reductions ranging from 10.6% to 37.6%, thereby demonstrating the cost-effectiveness of blending cloud resources for workflow execution. Note to Practitioners-Workflow scheduling in cloud environments is of significant practical relevance and has drawn considerable interest from both industry and academia. Blending serverful and serverless cloud resources offers a promising path toward cost-effective workflow execution, yet it introduces additional challenges in resource coordination, performance variability, and budget control. This work presents a hybrid resource management framework together with a budget-constrained workflow scheduling algorithm designed specifically for blended cloud environments. The proposed approach has been validated through real-world experiments and further assessed via comprehensive simulation studies. Results demonstrate that it consistently outperforms state-of-the-art methods in terms of makespan under user-defined budget limits. The proposed solution is practical and readily applicable in real cloud settings. It can assist cloud users and operators in efficiently allocating workflow tasks across heterogeneous cloud resources to achieve better performance within a specified budget.

Accurate and continuous outdoor pedestrian positioning using smartphones remains challenging in complex environments like urban canyons, where Global Navigation Satellite System (GNSS) signals are frequently degraded or blocked, and Pedestrian Dead Reckoning (PDR) suffers from cumulative errors. To address this, this paper proposes a novel fusion method based on a Robust Adaptive Cubature Kalman Filter (RACKF). The core of our approach is a two-stage filtering architecture: the first stage employs a quaternion-based RACKF to optimally fuse gyroscope and magnetometer data for robust heading estimation; the second stage performs the core fusion of GNSS observations with an enhanced 3D PDR solution. Key innovations include an adaptive noise estimation strategy combining fading and limited memory weighting, a robust M-estimator-based mechanism to suppress outliers, and the integration of differential barometric height measurements. Experimental results demonstrate that the proposed method achieves a horizontal positioning accuracy of 3.28 m (RMSE), outperforming standalone GNSS and improving 3D PDR by 25.97% and 10.39%, respectively. This work provides a practical, infrastructure-free solution for robust smartphone-based outdoor navigation.